词条 | Onsager–Machlup function |
释义 |
The Onsager–Machlup function is a function that summarizes the dynamics of a continuous stochastic process. It is used to define a probability density for a stochastic process, and it is similar to the Lagrangian of a dynamical system. It is named after Lars Onsager and S. Machlup who were the first to consider such probability densities.[1] The dynamics of a continuous stochastic process {{mvar|X}} from time {{math|t {{=}} 0}} to {{math|t {{=}} T}} in one dimension, satisfying a stochastic differential equation where {{mvar|W}} is a Wiener process, can in approximation be described by the probability density function of its value {{math|xi}} at a finite number of points in time {{math|ti}}: where and {{math|Δti {{=}} ti+1 − ti > 0}}, {{math|t1 {{=}} 0}} and {{math|tn {{=}} T}}. A similar approximation is possible for processes in higher dimensions. The approximation is more accurate for smaller time step sizes {{math|Δti}}, but in the limit {{math|Δti → 0}} the probability density function becomes ill defined, one reason being that the product of terms diverges to infinity. In order to nevertheless define a density for the continuous stochastic process {{mvar|X}}, ratios of probabilities of {{mvar|X}} lying within a small distance {{mvar|ε}} from smooth curves {{math|φ1}} and {{math|φ2}} are considered:[2]as {{math|ε → 0}}, where {{mvar|L}} is the Onsager–Machlup function. DefinitionConsider a {{mvar|d}}-dimensional Riemannian manifold {{mvar|M}} and a diffusion process {{math|X {{=}} {Xt : 0 ≤ t ≤ T} }} on {{mvar|M}} with infinitesimal generator {{math|{{sfrac|1|2}}ΔM + b}}, where {{math|ΔM}} is the Laplace–Beltrami operator and {{mvar|b}} is a vector field. For any two smooth curves {{math|φ1, φ2 : [0, T] → M}}, where {{mvar|ρ}} is the Riemannian distance, denote the first derivatives of {{math|φ1, φ2}}, and {{mvar|L}} is called the Onsager–Machlup function. The Onsager–Machlup function is given by[3][4][5] where {{math|{{!!}} ⋅ {{!!}}x}} is the Riemannian norm in the tangent space {{math|Tx(M)}} at {{mvar|x}}, {{math|div b(x)}} is the divergence of {{mvar|b}} at {{mvar|x}}, and {{math|R(x)}} is the scalar curvature at {{mvar|x}}. ExamplesThe following examples give explicit expressions for the Onsager–Machlup function of a continuous stochastic processes. Wiener process on the real lineThe Onsager–Machlup function of a Wiener process on the real line {{math|R}} is given by[6] [Proof] Let {{math|X {{=}} {Xt : 0 ≤ t ≤ T} }} be a Wiener process on {{math|R}} and let {{math|φ : [0, T] → R}} be a twice differentiable curve such that {{math|φ(0) {{=}} X0}}. Define another process {{math|Xφ {{=}} {Xtφ : 0 ≤ t ≤ T} }} by {{math|Xtφ {{=}} Xt − φ(t)}} and a measure {{math|Pφ}} by For every {{math|ε > 0}}, the probability that {{math|{{!}}Xt − φ(t){{!}} ≤ ε}} for every {{math|t ∈ [0, T]}} satisfies By Girsanov's theorem, the distribution of {{math|Xφ}} under {{math|Pφ}} equals the distribution of {{mvar|X}} under {{mvar|P}}, hence the latter can be substituted by the former: By Itō's lemma it holds that where is the second derivative of {{mvar|φ}}, and so this term is of order {{mvar|ε}} on the event where {{math|{{!}}Xt{{!}} ≤ ε}} for every {{math|t ∈ [0, T]}} and will disappear in the limit {{math|ε → 0}}, hence Diffusion processes with constant diffusion coefficient on Euclidean spaceThe Onsager–Machlup function in the one-dimensional case with constant diffusion coefficient {{mvar|σ}} is given by[7] In the {{mvar|d}}-dimensional case, with {{mvar|σ}} equal to the unit matrix, it is given by[8] where {{math|{{!!}} ⋅ {{!!}}}} is the Euclidean norm and GeneralizationsGeneralizations have been obtained by weakening the differentiability condition on the curve {{mvar|φ}}.[9] Rather than taking the maximum distance between the stochastic process and the curve over a time interval, other conditions have been considered such as distances based on completely convex norms[10] and Hölder, Besov and Sobolev type norms.[11] ApplicationsThe Onsager–Machlup function can be used for purposes of reweighting and sampling trajectories,[12] as well as for determining the most probable trajectory of a diffusion process.[13][14] See also
References1. ^Onsager, L. and Machlup, S. (1953) 2. ^Stratonovich, R. (1971) 3. ^Takahashi, Y. and Watanabe, S. (1980) 4. ^Fujita, T. and Kotani, S. (1982) 5. ^Wittich, Olaf 6. ^Ikeda, N. and Watanabe, S. (1980), Chapter VI, Section 9 7. ^Dürr, D. and Bach, A. (1978) 8. ^Ikeda, N. and Watanabe, S. (1980), Chapter VI, Section 9 9. ^Zeitouni, O. (1989) 10. ^Shepp, L. and Zeitouni, O. (1993) 11. ^Capitaine, M. (1995) 12. ^Adib, A.B. (2008). 13. ^Adib, A.B. (2008). 14. ^Dürr, D. and Bach, A. (1978). Bibliography{{refbegin|30em}}
External links
3 : Functional analysis|Functions and mappings|Stochastic processes |
随便看 |
|
开放百科全书收录14589846条英语、德语、日语等多语种百科知识,基本涵盖了大多数领域的百科知识,是一部内容自由、开放的电子版国际百科全书。